AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

RMP-Loss: Regularizing Membrane Potential Distribution for Spiking Neural Networks

Yufei Guo, Xiaode Liu, Yuanpei Chen, Liwen Zhang, Weihang Peng, Yuhan Zhang, Xuhui Huang, Zhe Ma

ICCV 2023 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 40 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add RMP-Loss regularizing membrane potential toward spike values (ablation)SNN ResNet20 · CIFAR-10 (T=4)accuracy: 91.26% -> 91.89%Improved
add RMP-Loss (ablation)SNN ResNet19 · CIFAR-10 (T=4)accuracy: 95.04% -> 95.51%Improved
add RMP-Loss (ablation)SNN VGG16 · CIFAR-10 (T=4)accuracy: 92.80% -> 93.33%Improved
add RMP-Loss (ablation)SNN ResNet20 · CIFAR-10 (T=4)quantization error: 0.186 -> 0.121Improved
RMP-Loss vs TET on ResNet19RMP-Loss · CIFAR-100 (T=4)accuracy: 74.47% -> 78.28%Improved
RMP-Loss vs InfLoR-SNN on ResNet19 at T=4 and T=6RMP-Loss · CIFAR-10accuracy: 96.27% vs 95.51% (T=4)Hurt
RMP-Loss vs InfLoR-SNN on CIFAR10-DVS ResNet19RMP-Loss · CIFAR10-DVS (T=10)accuracy: 75.50% -> 76.20%Improved
RMP-Loss vs integer-spike SEW ResNet on ImageNetRMP-Loss · ImageNet (ResNet34)top-1 accuracy: 67.04% vs 65.17%Hurt

Later papers that compared with this method

1 later accepted paper reports RMP-Loss as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy